CVE-2026-106513 in MISP
Summary
by MITRE • 10/06/2026
MISP exposes critical infrastructure settings—specifically the Redis host addresses used by the core application, the ZeroMQ plugin, and the SimpleBackgroundJobs plugin—through its web UI and API to site-admin users. The background job workers trust raw Redis job payloads without additional validation. An attacker who obtains a hijacked site-admin session (for example, through a stored cross-site scripting vulnerability) can modify the Redis host settings to point at an attacker-controlled Redis server and then restart the workers. Once the workers connect to the attacker's Redis instance, the attacker can inject malicious job payloads that the workers execute, achieving arbitrary command execution as the worker account. Additionally, the download_attachments_on_load setting, which controls inline attachment rendering, was modifiable through the same interface, allowing a hijacked session to re-enable a feature that could facilitate further client-side attacks. The vulnerability requires site-admin privileges and a prior session-compromise mechanism; it does not require unauthenticated access. The impact is remote code execution in the context of the MISP worker process and potential data exfiltration through the attacker-controlled Redis connection.
If you want to get best quality of vulnerability data, you may have to visit VulDB.
Analysis
by VulDB Data Team • 10/07/2026
The disclosed vulnerability represents a critical configuration manipulation flaw within the Malware Information Sharing Platform, specifically affecting how site administrators interact with backend infrastructure components such as Redis, ZeroMQ, and SimpleBackgroundJobs. The core issue stems from the platform's design choice to expose sensitive operational settings through its web user interface and application programming interface without implementing sufficient validation or integrity checks on the destination hosts defined by these configurations. This architectural oversight allows any authenticated site-administrator-level account to alter fundamental connection parameters that dictate how background job workers communicate with message brokers. The severity of this flaw is amplified by the fact that the underlying worker processes operate under a trust model that assumes configuration values are legitimate and safe, thereby failing to validate the origin or integrity of Redis host addresses before establishing connections.
The technical mechanism of exploitation relies on an attacker first gaining control over a site-administrator session through secondary vectors such as stored cross-site scripting vulnerabilities or other authentication bypass techniques. Once this elevated privilege is obtained, the adversary can navigate to the administrative settings interface and modify the Redis host configuration fields to point toward a maliciously controlled Redis server instance hosted by the attacker. Following this modification, the attacker triggers a restart of the background job workers through the API or UI. Upon restarting, these worker processes initialize their connections using the newly configured, hostile endpoint rather than the legitimate internal infrastructure. This shift in network topology effectively turns the MISP backend into an intermediary that accepts and relays commands from an external adversary without any cryptographic verification or source authentication of the configuration data itself.
The operational impact is severe, resulting in remote code execution within the context of the user account running the MISP worker processes. Because Redis job payloads are trusted raw inputs by these workers, once a connection to the attacker-controlled server is established, the adversary can inject maliciously crafted job definitions that exploit deserialization vulnerabilities or command injection flaws inherent in how the workers process incoming tasks. This leads directly to arbitrary code execution on the host system where MISP services reside. Furthermore, this configuration change facilitates potential data exfiltration, as the worker processes may inadvertently transmit sensitive internal data to the attacker's Redis instance during normal operation or while processing malicious jobs designed to leak information. The vulnerability also allows for the re-enabling of the download_attachments_on_load setting, which controls inline attachment rendering in email clients and web interfaces, thereby creating additional attack surfaces for client-side exploits against users viewing MISP content.
From a threat modeling perspective, this flaw aligns with CWE-20 Improper Input Validation regarding configuration parameters and CWE-94 Improper Control of Generation of Code (Code Injection) due to the execution of attacker-supplied payloads via trusted workers. In terms of MITRE ATT&CK framework mapping, the initial access phase involves hijacking a privileged session, while the exploitation phase corresponds to T1059 Command and Scripting Interpreter through the injection of malicious job tasks that are executed by system services. The lateral movement potential is limited but present if the worker account has broader network permissions or access to other internal systems via the Redis connection.
Mitigation strategies must focus on immediate remediation of configuration management practices within MISP deployments. Administrators should immediately rotate all site-administrator credentials and revoke any potentially compromised sessions to prevent further unauthorized changes. It is critical to audit recent modifications to infrastructure settings, specifically checking for any alterations to Redis host addresses or ZeroMQ endpoints that do not correspond to known administrative actions. Security teams should also review access logs for unusual API calls related to configuration updates during the period of suspected compromise. Long-term fixes require implementing strict input validation and allow-listing mechanisms for all critical infrastructure configuration fields in both the web UI and API layers. Additionally, introducing cryptographic signing or integrity checks for configuration payloads before they are applied by worker processes would prevent unauthorized modifications from taking effect even if an attacker gains administrative access. Restricting network egress rules to ensure that MISP workers can only connect to whitelisted Redis instances provides a robust defense-in-depth measure against this type of infrastructure hijacking.